The beneficial effects of body fat and adipose tissue in

International Journal of Obesity (1997) 21, 738±746
ß 1997 Stockton Press All rights reserved 0307±0565/97 $12.00
Mini-Review
The bene®cial effects of body fat and adipose
tissue in humans*
NG Norgan
Department of Human Sciences, Loughborough University, Loughborough, Leicester LE11 3TU, UK
Body fat and adipose tissue are considered to have bene®cial effects when they promote or protect the present and
future function. These effects do not occur at absolute amounts or percentages of the body weight but rather they are
context speci®c. Body fat stores are the major energy stores of the body and are important determinants of survival in
starvation or undernutrition. Reproduction features highly as a biological function. Humans are alone in having major
sex-speci®c fat stores and patterns of fat distribution1 and these have been linked with the onset and maintenance of
menstruation, with mate selection and sexual signalling, and with favourable pregnancy and lactation outcomes. To
survive and reproduce good physical and psychological health are advantageous attributes. Work metabolism, bone
health and, possibly immune function and energy balance itself, are related in functionally bene®cial ways to fat
content and distribution.
Keywords: body fat; survival; reproduction; mate selection; work capacity; bone health
Introduction
In addressing and dealing with the problems caused
by overweight and obesity, we tend to lose sight of the
fact that the human body evolved with a capacity to
store dietary energy as fat and that this ability has had
a selective advantage. The Association for the Study
of Obesity Workshop on `Good Fat, Bad Fat'Ð
clinical and metabolic aspects of adipose tissue distribution in November 1996 provided an opportunity
for a reminder of some bene®cial aspects of body fat
and fatness. Body fat and adipose tissue are considered to have bene®cial effects when they promote or
protect present and future function. These effects do
not occur at absolute amounts or percentages of the
body weight but rather they are context speci®c. What
in one set of circumstances may be bene®cial may be
dysfunctional in others.
Human fat contents
Energy is the key to life and, as intermittent feeders,
we require stores of energy and mechanisms to
*This review is based on a presentation given at the meeting on
`Good Fat, Bad Fat' held in London by the Association for the
Study of Obesity in November, 1996.
Received 25 March 1997; revised/accepted 12 May 1997
accumulate and release these stores in the short and
long term. Fat, here taken to be triacylglyceride, is
quantitatively the most important form of stored
energy and is found in adipose tissue (AT). There is
a sexual dimorphism in fat content and distribution.
Typical fat contents are 10 and 15 kg, 15 and 27% of
body weight, as in the Reference Man and Woman of
Albert Behnke.2 These are not averages or desirable
amounts but rather models based on average dimensions to be used as comparators for other data. These
amounts have been separated into essential and
storage fat; 3 and 12% of body weight in men and 9
and 15% in women, respectively. Essential fat is that
in bone marrow, heart, lungs, liver, kidneys, intestines, muscles and lipid-rich tissue of the central
nervous system with roles other than energy storage.
It includes other fats in addition to triacylglycerides,
such as phospholipids. In the model of Reference
Woman, the essential fat also includes sex speci®c
fat, primarily in the mammary glands, the pelvic
region and the thighs. There is no general agreement
of the quantities of these and whether they are strictly
essential or a storage reserve for the energy demands
of pregnancy and lactation. Katch and colleagues3
have suggested subdividing storage fat so that there
are three fat components in women, essential (4%),
sex-speci®c reserve storage (5%) and expendable
storage (15%), on average.
The sexual dimorphism in adult fat distribution is
particularly marked. Men have a central pattern of
deposition, mainly on the trunk and abdomen. In
Bene®cial fat
NG Norgan
women deposition is, characteristically, on the buttocks and thighs. The sex-speci®c amounts and distribution, particularly the regional adiposity in the
abdominal and gluteo-femoral regions, are caused, in
a large part, by the sex hormones. Testosterone may
facilitate deposition in the former and utilisation in the
latter and the reverse for oestrogen. According to
Martin and Daniel4 the effects are more related to
the oestrogen/androgen balance than to oestrogen
alone. They consider that the suggestion that sex
hormones have a major role in sex differences has
not been proved. The enzyme lipoprotein lipase (LPL)
facilitates fat uptake and storage in adipose tissue. Its
level is high in the adipose tissue at the hip, thigh and
breast in women and in the abdomen in men. Variations in LPL levels may contribute to differences in fat
distributions between the sexes, between individuals
of the same sex, between ethnic groups, during
pregnancy and in middle age.
The responses of adipose tissue to metabolic and
nutritional change vary at different sites. Truncal
deposits appear to be the primary energy stores.
Even at the same subcutaneous site super®cial layers
seem less responsive perhaps because they have more
of a structural role.5
Views on the lowest body mass index (BMI)
compatible with life have seen downward revisions
in recent years. James et al11 proposed a BMI of 12 as
the absolute lower limit compatible with life. Henry8
in a more detailed review endorsed this proposal but
thought an apparent sex difference of 13 in males and
11 in females warranted further investigation. It is
now clear from data collected in famine victims that
values as low as 10 are compatible with life with
specialised care.10,12
A second effect of body fat on survival, in addition
to the size of the energy stores, is that the types of
responses to energy de®cits are related to the initial
level of energy stores in the body, at least in studies of
more than four weeks duration.13 The nitrogen (N) or
protein loss during starvation or undernutrition is less
in individuals with high fat contents than those of low
fat contents and less in females than in males. It is
postulated to be constant in an individual during both
storage and mobilisation until the preterminal phase of
starvation.14 It appears as though the thin are eking
out their fat stores in a protective response but the
greater N or protein losses appears maladaptive.
Hazards of thinness
Survival
Energy stores and energy ¯ux
Typical healthy young men and women with body fat
stores of 15 and 25% of body weight, some 10 and
15 kg of fat, have fat energy stores of 370 and 555 MJ,
respectively. At an energy expenditure of 6 MJ/d,
these would provide suf®cient energy for 60±90 d.
There is also energy available from body protein but
only half of the protein, containing some 100 MJ, can
be mobilised without becoming life threatening. The
arithmetic seems simple and compelling. The larger
the fat stores when entering a period of starvation or
chronic energy de®ciency, the longer the survival
period or the less threatening it is. At death, body
weight loss may be as much as 50% by which time all
fat stores will be depleted. The differential survival
rates of males and females under conditions of dietary
energy stress are well known from studies in war-time
Europe and elsewhere and attest to the importance of
energy stores in starvation.6±10
The arithmetic is not, however, uncomplicated
because of the adaptations to negative energy balance
that occur. Weight, fat and energy are not lost at a
constant rate throughout a period of undernutrition or
starvation. Firstly, a smaller body size lowers the
energy requirement. Secondly, volitional physical
activity falls, further lowering the energy requirement.
There may also be a reduction in metabolic rate per
unit of non-osseous lean tissue. Thus, on a constant,
initially inadequate energy intake, the rate of weight
and fat loss decreases exponentially.
There has been considerable debate over the health
risks of low weight and BMI in young adults and the
middle-aged. The U or J-shaped relationships between
mortality or probability of death and low weight or
BMI have been interpreted as the effect of the inclusion of smokers and the already ill causing the lefthand tails. A recent meta-analysis by Troiano et al15
considered the 19 major prospective cohort studies
that met strict inclusion criteria. They concluded that
increased mortality persisted at moderately low BMI
in white men after smoking or existing disease were
accounted for. It was comparable to that observed at
extreme overweight. The limited data on women
indicated that there was little relationship between
BMI and mortality in non-smokers over a 10 y period
beginning 50 y of age. The 16 y follow-up of 115 000
30±55 y old US women in the Nurses' Health Study16
also showed no increase in mortality in leaner women
who had never smoked. Is this sex difference in
mortality experience evidence of a fat content
effect? It certainly demonstrates the hazards of thinness in men. The American Health Foundation's
Expert Panel on Healthy Weight,17 in a departure
from other weight recommendations, eschewed publishing upper and lower limits of weight to avoid
people losing more than may be necessary to achieve
improved health.
Dietary restriction is well known to increase longevity in laboratory animals. Initially, the mechanism
was thought to be that excess fat caused premature
death. Now it appears that in dietary restricted animals
longevity and body fat content are positively correlated.18 This observation does not ®t easily with any of
739
Bene®cial fat
NG Norgan
740
the three current hypotheses of the mechanism by
which dietary restriction increases longevity.
Lower limit of body fat in healthy men
Humans can survive with tiny amounts of body fat as
can be seen in body builders and wrestlers, who both
have normal or increased amounts of skeletal muscle,
and in anorectics and famine victims. Friedl and
colleagues9 addressed the question of what is the
lower limit of body fat in healthy adults by studying
55 young men undergoing strenuous US Army leadership training with food de®cits. There was a 5 MJ/d
energy de®cit over eight weeks, leading to 16%
weight loss. This level of food restriction approached
the voluntarily tolerable limit for these highly motivated men. Body composition was measured by
DEXA technology, thought to be less affected by
changing compositions, although DEXA did overestimate gravimetrically determined body weight by
2±3 kg at six and eight weeks. When minimum body
fat percentages of 4±6% were reached, some 2.5±4 kg
of fat, increasing amounts of lean tissue were sacri®ced. This was demonstrated by continuing decreases
in waist, hip and thigh circumferences when body fat
was low and had stopped falling. This minimum range
compares well with the data at the end of the Minnesota study of 5.3% body fat, allowing for changing
fat-free mass composition, and Behnke's Reference
man with 3% or 2.1 kg of essential fat. Any encroachment into this impairs normal physiological function
or capacity for endurance exercise. Although low
energy levels may never be a cause of death they
are a major contributory factor in the breakdown of
adaptation and maintenance of homeostasis and the
impairment of physiological function. Energy stores,
particularly fat stores, protect function.
The role of body fat in insulation and survival in cold
air or water
A classical environmental physiology dictum is that
subcutaneous and deep body fat are major components determining the rate of internal heat transfer
during water immersion. The bene®ts of body fat in
the maintenance of core temperature during cold
water immersion have been amply and consistently
veri®ed.19 In addition to the insulative effects, women
derive considerable hydrodynamic bene®ts from their
greater fat content, which increases buoyancy, and the
greater distribution of fat on the legs, making them
more horizontal in the water with less drag. The
record time for swimming the English Channel was
held by a woman between 1978 and 1994.
Body fat, particularly subcutaneous fat, also provides signi®cant protection during cold air exposure.
However, Inuits (Eskimos) that followed a traditional
way of life before acculturation had low skinfold
thicknesses and Korean diving women who made a
living by regularly diving in cold water had less
subcutaneous fat than other Korean women and
much less than American women.20 Pond21 has
found that polar bears have proportions of fat situated
subcutaneously not signi®cantly different to those of
temperate zone and tropical carnivores. This suggests
that other physiological mechanisms such as vasoconstriction or counter-current circulation maybe as
important as insulation in heat retention. Alternatively, the role of non-active, minimally perfused
skeletal muscle acting as an insulator needs to be
considered. The most commonly observed adjustment
by circumpolar residents is a blunted shivering
response.22
Reproduction
In evolutionary terms, survival itself is of little consequence. It is only advantageous for reproduction and
for protecting the genes of the next generation. It has
been suggested that because human reproduction is
such an energy demanding process where the chances
of a successful outcome are low mechanisms may
have evolved to restrict reproduction at the outset.
Fertility
As is well known, Frisch and co-workers23 suggested
a critical body weight of 46±47 kg for menarche in
American and European girls, later translated to 17%
body fat for menarche and 22% for regular menstrual
cycles.24 Many objections have been raised about
these limits and the treatment of the data in arriving
at them, particularly the assessment of % fat from
height and weight, and the hypothesis has been
modi®ed and extended. Famine and anorexia nervosa
are known to affect reproductive function but attempts
to apply the hypothesis to more moderate nutritional
challenges have not been successful, except for
physically active individuals particularly endurance
runner and ballet dancers. Dif®culties remain, such as
observations that the % fat of menstruating and nonmenstruating runners have been found to be similar25
and many individuals do not behave according to the
thresholds. The proposed amounts of stored energy
required are very large compared with other species
where pregnancy, often involving multiple births, and
lactation are more energetically expensive. Brownell
et al26 suggested that regional fat loss rather than total
fat loss may trigger amenorrhea. Deletion of fat on
hips, thighs and buttocks, which provide much of the
energy for pregnancy and lactation, may disrupt
reproductive function. DeRidder et al27 found that
weight matched girls with lower waist-to-hip ratios
(WHR) had early pubertal endocrine activity than
those with higher WHR.
Frisch28,29 has suggested a number of ways in
which adipose tissue may regulate female reproduction through the hypothalamic-pituitary-ovarian axis.
Firstly, adipose tissue of the breast and abdomen
Bene®cial fat
NG Norgan
is a signi®cant extra-gonadal source of oestrogen,
accounting for a third of the circulating oestrogen in
pre-menopausal women. Secondly, fatness seems to
in¯uence the direction of oestrogen metabolism to
more potent forms in the obese or less potent forms in
the lean. Thirdly, those obese women and young girls
who are relatively fatter have a diminished capacity
for oestrogen to bind to serum sex hormone-binding
globulin (SHBG), leading to higher concentrations of
the free form. SHBG also regulates the availability of
oestradiol to the brain and other target tissues. Finally,
the adipose tissue of obese women is known to store
oestrogens. The question remains as to whether these
effects in¯uence ovarian function.
Many other factors are now being implicated in the
regulation of the hypothalamic-pituitary-ovarian axis
and most workers consider physical and psychological
stress leading to increased adrenal corticosteroid and
catecholamine excretions, energy drain, and nutritional inadequacies to be equally if not more important than low adipose tissue masses. There is a
complex interplay of many factors on menstrual
function. It could be that low fat levels re¯ect the
discipline of dietary restriction and training rather
than having a causative effect. Attention is switching
away from athletes, with physiques that could have a
bias towards lower ovarian activity, and the presence
or absence of menstruation towards the more subtle
effects of hormone levels, such as the shortened luteal
phase of the cycle, the appearance of anovulatory
cycles and, in particular, the down-regulation of the
hypothalamic gonadotrophin pulse generator.30
Human mate selection: the role of shape preferences
Darwinian theory has not been applied successfully to
human affairs. The previous attempts by Herbert
Spencer at the end of the 19th century, the application
of neo-Darwinism to the social sciences in the 1920s
and the efforts of sociobiology in the 1970s failed,
perhaps because of the emphasis on rigid behaviours
rather than inbuilt behavioural predispositions highly
attuned to environmental factors. The latest attempt at
application is occurring under the name of `evolutionary psychology'. One example is in the area of
human mate selection. Men and women select mating
partners who will enhance their reproductive success,
according to evolutionary theories of human mate
selection. For women, this may be high status men
who control resources. For men, reproductive success
may be increased by choosing a woman with the
characteristics of being fecund and a good mother.
Singh31 states that a fundamental assumption of all
evolutionary theories of human mate selection is that
physical attractiveness is largely a re¯ection of reliable cues to a woman's reproductive success. The
assumption is that men assign great signi®cance to
good looks and this is said to be a cross-cultural
universal. It is said to be easy to identify high status
men but fecundity is not overt and more dif®cult to
identify. In the absence of direct signals of ovulation
or fertility in humans, in contrast to the non-human
primates, indirect cues are said to be used.
Singh31 asserts that the female hour-glass body
shape distinguishes menstruating from pre-menarchal,
pregnant and post-menopausal women and WHR is
the best indirect index of this. He showed that rankings of the attributes of `healthy', `youthful-looking
women', `attractive and sexy', and `desire and capacity for having children' were highest in the lowest
WHR shapes presented to rankers, irrespective of the
shapes being of underweight (BMI ˆ 15), normal
(BMI ˆ 20) or overweight (BMI ˆ 25). Fatness and,
surprisingly, thinness were perceived as unattractive.
Singh32 found no evidence that young black men and
women in the USA behaved differently by regarding
overweight female ®gures as desirable and attractive.
Physicians of both sexes also assigned higher rankings
of health, reproductive capability, and attractiveness,
to normal weight, low WHR female outlines.32 For a
trait to be considered adaptive it must be transgenerationally stable and Singh reported ratings of attractiveness and other attributes to be similar in all age
groups.31
It is necessary to establish a link between attributes
of attractiveness and physiological mechanisms regulating components of ®tness such as health, fecundity, successful pregnancy and nursing. Body fat and its
distribution may provide the link, with the high levels
of circulating oestrogen in women producing a lower
WHR. Singh31 notes that in the work of De Ridder27
on weight matched girls, those with lower WHR had
early pubertal endocrine activity and married women
with higher WHR and lower BMI are reported to have
more dif®culty getting pregnant and to have the ®rst
birth at later age.34 Also, in a Dutch prospective study
on arti®cial insemination, a 0.1 unit increase in WHR
led to a 30% decrease in probability of conception,
after controlling for many interfering variables.35
The conclusion from Singh's studies on Western
populations is that body fat and its distribution plays a
crucial role in judgements of female attractiveness,
health, youthfulness, and reproductive potential. The
suggestion is not that shape as evinced by WHR is the
only factor in mate selection but one involved in the
initial stages, `a ®rst wide pass ®lter'. At another
level, cultural factors come into play, as exempli®ed
by the different socio-economic status and obesity
relationships in the First and Third Worlds.
How does Singh's theory of the role of shape
preferences in human mate selection ®t in with the
supposed secular changes in de®nitions and preferences of attractiveness, particularly the shift from the
hour glass shape to the tubular? Studies of Playboy
centrefolds36 and Miss America competitors37 suggest
that the hour glass shape, and the WHR in particular,
have been maintained, even though body weights and
bust sizes of participants have decreased. Currently,
the curvy shape is regaining popularity as a glamorous
attribute. It is quite possible that it is features that are
741
Bene®cial fat
NG Norgan
742
rare or dif®cult to achieve that are sought. The history
of fashion shows us that during times of scarcity skirts
with plenty of material were fashionable. In times of
plenty, skirts are short and narrow. Likewise, body fat
is socially desirable and esteemed in times or areas of
shortage.38 Whether thinness or fatness is dif®cult to
achieve the attraction of the proportionally wider hip
than waist remains.
Waists and buttocks are uniquely human features,
not shared by the great apes. The evolution of human
bipedal locomotion has been associated with changes
to the mechanisms of pelvis and femur. The buttocks
arise from the rotation of the sacrum and pelvis and
the development of the gluteus maximus for bipedal
posture. The waist has its origins in the short ilium,
the development of the lumbar vertebrae and the
lumbar curve of the spine and the subcutaneous
adipose tissue on the gluteal facial. Pond39 has speculated that the WHR effect could be a vestigial form of
selection for correct pelvis shape, in that it signalled
the erect posture of the female early hominid and her
ability to give birth to large-brained offspring. The
locomotory in¯uences on body shape also allows male
attractiveness to have a role in mate selection.
Whereas men imaging women admire a muscular
chest, arms and shoulders, they actually nominate
the buttocks as the most attractive part of a man.40
Do these signal endurance capacity and hunting
prowess?
Pregnancy and lactation
There are a number of sex and regional differences in
adipose tissue content and metabolism.41 The sexes
differ most noticeably in their gluteofemoral fat content. The gluteofemoral fat cells are bigger than other
subcutaneous adipose tissue cells. They have
increased lipoprotein lipase levels, a rate limit on fat
uptake, compared with mammary and abdominal
cells, and a lower lipolytic activity due to more
alpha than beta adrenergic activity. Therefore, they
tend to accumulate fat. This pattern is accentuated
during pregnancy but disappears during lactation. As a
consequence, gluteofemoral fat deposits are regarded
as deposits to meet the energy demands of lactation.
After the menopause, these regional differences disappear although this can be reversed with hormone
replacement therapy.
Pregnancy
Fat stores are bene®cial for successful pregnancy
outcome. Low maternal weight before pregnancy
and poor weight gain during pregnancy are known
to result in an increased prevalence of low birthweight
infants.42 van der Spuy et al43 studied the outcome of
pregnancy and undernutrition in 1212 spontaneously
ovulating women and in 41 women in whom ovula-
tion had been induced to treat anovulatory infertility.
They found that in spontaneously ovulating women
those with a BMI less than 19 (n ˆ 179) had nearly
double the expected rate of low birth weight babies
(less than the 10th percentile of reference data). Birth
weight was corrected for sex, parity, gestational age
and maternal height. Women in whom ovulation was
induced had higher risk of low birth weight babies
(25% of births) increasing to 54% if they were underweight. van der Spuy43 concluded that the most
suitable treatment for infertility secondary to weight
related amenorrhea is dietary rather than induction of
ovulation.
Lactation
Lactation is an energetically expensive process and
the fat stores laid down in pregnancy are often
regarded as stores for lactation. In women in the
UK, however, the sources of energy for milk production have been found to be from increased energy
intake (62%), from decreased physical activity (35%)
and from decreased BMR (3%).44 There may also be a
reduction in diet-induced thermogenesis. In women in
developing countries, there may be less scope for
increasing energy intakes and decreasing activity
and fat stores may be more important to them. Prentice et al 45 reviewed the world literature of 41
databases to test whether relationships existed
between low BMI and breast milk quantity and
quality. Milk production was not affected by low
BMI nor was composition with the possible exception
of milk fat levels. Human lactation performance therefore appears very robust and unaffected by low energy
stores. To achieve this, energy intakes must be raised
so this robustness may not persist in famine or near
famine conditions and fat energy stores will prove
bene®cial.
Some metabolic aspects
Many metabolic aspects of adipose tissue are discussed fully in other reviews arising from the ASO
Workshop on `Good Fat, Bad Fat'. Because of the
interest in obesity, most attention has been focused on
the large adipose tissue depots and their relation to
health. However, there are suggested roles for some of
the smaller discreet adipose tissue deposits in protecting and promoting function. One such unique type is
brown adipose tissue.
Brown adipose tissue (BAT)
BAT, like many other tissues can oxidise a variety of
substrates but it is unique in that oxidation can be
uncoupled from oxidative phosphorylation, that is
substrate energy can be transduced directly to heat.
Uncoupling is under the control of the sympathetic
Bene®cial fat
NG Norgan
nervous system which also has a co-ordinating role by
increasing lipolysis and blood ¯ow with the effect of
transporting more oxygen to and heat from the tissue.
This has two possible advantageous outcomes, in
maintaining body temperature in the cold and in
regulating body energy content, as proposed by
Rothwell and Stock.46 Such regulatory adipose tissue
should be regarded as protective adipose tissue but its
signi®cance in adult humans is uncertain.
Immune function
Pond has proposed a role for adipose tissue in the
function of the immune system, noting that many
features of mammalian adipose tissue cannot be
explained as adaptations to energy storage, thermal
insulation or protection.47 Adipose tissue and the
cellular immune systems are partitioned into numerous small entities with major differences in the capacities for interaction. Pond concludes that immediate,
controllable access to lipolytic products of appropriate
composition is probably much more important for
local interaction with lymphoid cells in nodes than
access to large quantities of lipids. Investigation of
these small local deposits may prove a more fruitful
approach.
On the other hand, Stallone48 reviewed the meagre
evidence relating immune function to fatness and
concluded that obesity is associated with impairments
in host defence mechanisms and that certain types of
weight reduction strategies produce further alterations
in immune responsiveness. He too concluded that
more work was needed.
Fat and physical performance
There is a widespread impression that body fat is
always detrimental to physical performance. There is
a strong negative relationship between body fat and
performance in most activities, such as running,
gymnastics, ®gure skating and ballet, but this is not
so in long distance swimming. Also, it has been
known since the 1930s that high fat diets impair
endurance performance.49 However, the fat in adipose
tissue, and the smaller amounts but equally important
intra-muscular fat, play important roles in the metabolism of exercise, a role that may be blocked by
inappropriate carbohydrate consumption with a consequent impairment of work capacity.
The factor limiting work capacity in hard work, at
75% VO2 max, or 40±60 kJ/min, is muscle glycogen
stores. Greater utilisation of fat conserves the muscle
glycogen and prolongs work time. It is in this way,
that caffeine may have ergogenic effects in high
intensity endurance exercise. Most of the energy for
low intensity fasting exercise such as walking is
derived from plasma fatty acids from body fat, the
concentration of which increases as exercise proceeds.50 However, fat uptake by muscle is rate limited
by lipoprotein lipase and as exercise intensity
increases up to 65% of maximal oxygen uptake,
intra-muscular fat stores make an important contribution. At 85% of maximal oxygen uptake and above,
muscle glycogen and plasma glucose are the dominant
substrates. Trained individuals have lower exercise
respiratory exchange ratios (RER) indicating
increased fat uptakes and greater fat use, particularly
intramuscular fat, thus sparing muscle glycogen.51 Fat
in the metabolic mixture, particularly in hard work, is
bene®cial.
There are reports that women, with their greater
adipose tissue masses, have lower RERs in exercise
than men52 and in this respect they resemble trained
individuals. However, because of the great heterogeneity of protocols adopted and subjects investigated
there is still some doubt about sex differences in
substrate utilisation during exercise.53 Costill et al54
found similar RERs in endurance runners of both
sexes at 70% VO2 max after 60 min exercise. Thus,
the differences between sexes diminishes as the level
of cardiorespiratory ®tness is increased to that of
highly trained individuals.53
In the context of a discussion of bene®cial effects of
fat, the question arises whether the sex differences in
responses to exercise are related to fat content? The
increased utilisation of fat in women compared with
men exercising at 60±75% VO2 max, was found not to
be proportional to fat weight, suggesting it is not a
direct effect.55
Women are reported to have more Type 1 muscle
®bres than men.41 Their ®bre diameters are smaller,
particularly those of fast twitch ®bres, giving rise to
larger Type 1 fractional volume. This could contribute
to a higher potential for oxidative metabolism per unit
muscle in women. In addition, a sex difference in the
adrenergic regulation of lipid mobilisation has been
described.56 In men, both beta and alpha adrenergic
receptors in adipose tissue are activated. In women,
only beta receptors are activated.
Bone health
Hip fractures are a major public health problem in
many parts of the world, involving high rates of
hospital bed occupancy and reduced quality of life
for the sufferers. Fracture risk depends on attributes of
bone, such as bone mineral density, the trabecular
(inner lacey bone) integrity and microfracture healing,
and the propensity to falls.
It is well established that bone mass is positively
associated with body weight. There has been debate as
to whether the effects were due to adipose tissue or
lean tissue or both. A recent international multicentre
study of the in¯uence of adipose and lean mass on
bone mineral content has provided evidence for differences in their relative in¯uences at the major
fracture sites.57 One thousand and six hundred early
postmenopausal women less than 130% ideal weight
from four areas in US and Europe were recruited
743
Bene®cial fat
NG Norgan
744
using population based strategies. The bone sites
investigated included the hip, the femoral neck, the
trochanter, spine and radius. Bone mineral content
(BMC) was measured by Hologic DEXA. The results
were expressed as percent differences over the interquartile ranges of weight, lean and adipose tissue,
determined by regression analysis controlling for
several interfering variables.
Body weight had strong associations at all skeletal
sites examined with BMC differences of 4±6% per
interquartile range. These seem small but they are
clinically signi®cant. For longitudinal studies, 5%
differences in BMC have been shown to correspond
to 17±24% changed fracture risk. In clinical trials, the
differential risk is double. The associations of BMC
with the fat and lean components were more variable,
lean 2±7%, fat 4±6% but low fat mass or low lean
mass could both adversely affect the major fracture
sites. The bones with the greatest differences in
fracture rates according to fat mass were the most
highly trabecular sites. The associations were similar
in direction and comparable in magnitude at each
geographic location.
The mechanisms whereby fat and lean masses
in¯uence bone mineral content may involve both
hormonal effects and physical loading. Conversion
of adrenal androgen to oestrogen occurs in adipose
tissue, androgen conversion being the most important
source of endogenous oestrogen in post menopausal
women. Low fat masses could therefore result in low
bone masses. The spine and distal radius are two of
the sites most affected, suggesting a whole body
effect. Lean mass makes up most of the body
weight and it may in¯uence bone by a loading
effect and by the mechanical stresses of activity and
muscle tension. The hip and neck and shaft of femur
were most in¯uenced by lean mass, suggesting independent regional effects.
The results of Hla et al57 agree with those of
Compston et al,58 Sowers et al 59 and Edelstein and
Barrett-Connor.60 Reid et al61 thought fat mass more
important than lean and neither oestrone production or
physical loading could explain the effect while
Sowers et al 59 concluded that higher fat mass is
only protective with higher lean mass.
The down-side of higher fat and lean masses is that
higher body weights will result in increased forces
across the joints which, with systemic factors, may
predispose to osteo-arthritis. There is a 4±5 fold
increase in risk of osteo-arthritis with BMI > 30
compared to < 25.62 The effect may be less at the
hip than the knee because the forces generated by
excess weight there are lower. However, obesity is
also related to osteo-arthritis in non-weight bearing
joints, such as the hand.
The protective effects of fat has been considered in
terms of functional protection but adipose tissue can
have a strictly mechanical protecting effect too. Thick
layers of subcutaneous adipose tissue afford protection to the limbs and internal organs from falls, blows
and other physical trauma. Adipose tissue deposits
within the orbits of the eye and on the palms and soles
also appear to have important protective functions as
they are much less affected by starvation than other
sites.
The risk of hip fracture from a fall is known to be
lower with higher BMI. Lauritzen63 in his review of
the pathogenic mechanisms leading to hip fracture
identi®es the soft tissue covering the hip as an
important determinant of hip fracture. Women with
hip fractures have a lower body weight compared with
controls, and they may also have less soft tissue
covering the hip, even when adjusted for BMI, indicating a more android body habitus. Experimental
studies show that the passive energy absorption in
soft tissue covering the hip may in¯uence the risk of
hip fracture and be an important determinant of hip
fracture perhaps even more important than bone
strength. Lauritzen developed and tested external hip
protectors in an open randomised nursing home study.
The rate of hip fracture was reduced by 50%, corresponding to 9 out of 247 residents saved from sustaining a hip fracture, a valuable outcome.
A human biologist's end note
Obesity is an extensive and increasing problem in
most parts of the world. The human body is a machine
of extraordinary design but why are there so many
¯aws in design that lead to disease? Why has natural
selection not selected for genes that, for instance,
control energy intake and/or expenditure better? One
obvious answer could be that a limit to fat stores has
never had an advantage in reproductive ®tness in the
past. But the body is best seen as a series of compromises. Most medical research seeks proximate
explanations to problems, the `what' and `how'.
Evolutionary explanations, the `why' have been
neglected, perhaps because of the dangers of teleology, ascribing purpose to natural phenomena. The
approach of Darwinian medicine is going some way to
offering explanations in terms of unrecognised
bene®ts to such apparently disadvantageous responses
as fever and pregnancy sickness.64 Equally, however,
natural selection does not create every adaptation that
would be valuable, for example monkeys with prehensile tails evolved in South America but not Africa.
From the evidence presented, in biological terms, in a
variety of conditions, much of the body fat of humans
can be considered to be bene®cial fat.
References
1 Pond CM. Morphological aspects and the ecological consequences of fat deposition in wild vertebrates. Ann Rev Ecol
Syst 1978; 9: 519±570.
2 McArdle WD, Katch FI, Katch VL. Exercise Physiology:
Energy, Nutrition and Human Performance. 3rd edn. Lea &
Febiger: Philadelphia, 1991, pp 600±604.
Bene®cial fat
NG Norgan
3 Katch VL, Campaigne B, Freedson P, Sady S, Katch FI,
Behnke AR. Contribution of breast volume to body fat
distribution in females. Am J Phys Anthropol 1980; 53: 93±
100.
4 Martin AD, Daniel M. Regional adiposity and the sex hormones. In: Duquet W, Day JAP (eds). Kinanthropometry IV.
E & FN Spon: London, 1993, pp 19±41.
5 Alexander HG, Dugdale AE. Fascial planes within subcutaneous fat in humans. Eur J Clin Nutr 1992; 46: 903±906.
6 Keys A, Brozek J, Henschel A, Mickelson O, Taylor HL. The
Biology of Human Starvation. University of Minnesota Press:
Minnesota, 1950.
7 Widdowson EM. The response of the sexes to nutritional
stress. Proc Nutr Soc 1976; 35: 175±180.
8 Henry CJK. Body mass index and the limits of human
survival. Eur J Clin Nutr 1990; 44: 329±335.
9 Friedl KE, Moore RJ, Martinez-Lopez LE, Askew EA, Marchitelli LJ, Hoyt RW, Gordon CC. Lower limit of body fat in
healthy active men. J Appl Physiol 1994; 77: 933±940.
10 Collins S. The limit of human adaptation to starvation. Nature
Med 1995; 1: 810±814.
11 James WPT, Ferro-Luzzi A, Waterlow JC. De®nition of
chronic energy de®ciency in adults. Report of a Working
Part of the International Dietary Energy Group. Eur J Clin
Nutr 1988; 42: 969±981.
12 Ferro-Luzzi A, James WPT. Adult malnutrition: simple
assessment techniques for use in emergencies. Br J Nutr
1996; 75: 3±10.
13 Forbes GB. Lean body massÐbody fat interrelationships in
humans. Nutr Rev 1987; 45: 225±231.
14 Payne PR, Dugdale AE. A model for the prediction of energy
balance and body weight. Ann Hum Biol 1977; 4: 525±535.
15 Troiano RP, Frongillo Jr EA, Sobal J, Levitsky DA. The
relationship between body weight and mortality: a quantitative
analysis of combined information of existing studies. Int J
Obes 1996; 20: 63±75.
16 Manson JE, Willett WC, Stampfer MJ, Colditz GA, Hunter
DJ, Hankinson SE, Hennekens CH, Speizer FE. Body weight
and mortality among women. New Engl J Med 1995; 333:
677±685.
17 Meisler JG, St Jeor S. Summary and recommendations from
the American Health Foundation's Expert Panel on Healthy
Weight. Am J Clin Nutr 1996; 63: 474S±477S.
18 Masoro EJ. McCay's hypothesis: undernutrition and longevity.
Proc Nutr Soc 1995; 54: 657±664.
19 Toner MM, McArdle WD.Human thermoregulatory responses
to acute cold stress with special reference to water immersion.
In: Fregly MJ, Blatteis CM (eds). Handbook of Physiology.
Section 4 Environmental Physiology. Vol. 1. Oxford University Press: New York, 1996.
20 Rennie DW, Covino BG, Howell BJ, Song SH, Kang BS,
Hong SK. Physical insulation of Korean diving women. J Appl
Physiol 1962; 17: 961±966.
21 Pond CM. An evolutionary and functional view of mammalian
adipose tissue. Proc Nutr Soc 1992; 51: 367±377.
22 Young AJ. Homeostatic responses to prolonged cold exposure:
human cold acclimatisation. In: Fregly MT, Blatteis CM (eds).
Handbook of Physiology. Section 4 Environmental Physiology.
Vol. 1. Oxford University Press: New York, 1996, pp 419±
438.
23 Frisch RE, Revelle R. Height and weight at menarche and the
initiation of the adolescent growth spurt in girls: estimated
total water, lean body weight and fat. Hum Biol 1971; 45: 469±
483.
24 Frisch RE, McArthur JW. Menstrual cycles: fatness as a
determinant of minimum weight for height necessary for
their maintenance or onset. Science 1974; 185: 949±951.
25 Katch FI, Spiak DL. Validity of the Mellits and Cheek method
for body-fat estimation in relation to menstrual cycle status in
athletes and non-athletes below 22 per cent fat. Ann Hum Biol
1984; 11: 389±396.
26 Brownell KD, Steen SN, Wilmore JH. Weight regulation
practices in athletes: analysis of metabolic and health effects.
Med Sci Sports Exer 1987; 19: 546±556.
27 DeRidder CM, Bruning PF, Zonderland ML, Thijssen JHH,
Bonfrer JMG, Blankstein MA, Huisveld JA, Ericj WBM. Body
fat mass, body fat distribution and plasma hormones in early
puberty in females. J Clin Endocrinol Metabolism 1990; 70:
888±893.
28 Frisch RE. Body fat, menarche, ®tness and fertility. In: Frisch
RE (ed). Adipose Tissue and Reproduction. Karger: Basel,
1990, pp 1±26.
29 Frisch RE. The right body weight: body fat, menarche and
fertility. Proc Nutr Soc 1994; 53: 113±129.
30 Lunn PG. Breast-feeding practices and other metabolic loads
affecting human reproduction. In: Rosetta L, Mascie-Taylor
CGN (eds). Variability in Human Fertility. Cambridge University Press: Cambridge, 1996, pp 195±216.
31 Singh D. Adaptive signi®cance of female physical attractiveness: role of waist-to-hip ratio. J Person Soc Psych 1993; 65:
293±307.
32 Singh D. Body fat distribution and perception of desirable
female body shape by young black men and women. Int J
Eating Dis 1994; 16: 289±294.
33 Singh D. Waist-to-hip ratio and judgement of attractiveness
and healthiness of female ®gures by male and female physicians. Int J Obes 1994; 18: 731±737.
34 Kay SA, Folsom AR, Prineas RJ, Potter JD, Gapstur SM. The
association of body fat distribution with lifestyle and reproductive factors in a population study of postmenopausal
women. Int J Obes 1990; 14: 583±591.
35 Zaadstra BM, Seidell JC, Van Noord PAH, te Velde ER,
Habbema JDF, Vrieswijk B, Karbatt J. Fat and female fecundity: prospective study of effect of body fat distribution on
conception rates. BMJ 1993; 306: 484±487.
36 Mazur A. U.S. trends in feminine beauty and over adaptation.
J Sex Res 1986; 22: 281±303.
37 Garner DM, Gar®nkel PE, Schwartz D, Thompson M. Cultural
expectations of thinness in women. Psych Rep 1980; 47: 183±
191.
38 de Garine I, Pollock NJ (eds). Social Aspects of Obesity.
Gordon and Breach Publishers: London, 1995.
39 Charter D. 0.85 seeks 0.7 for 2.4. Times Higher Education
Supplement 11, August, 1995, p 15.
40 Wilson G, Nias D. Love's Mysteries: the Psychology of Sexual
Attraction. Open Books: London, 1976, pp 23±25.
41 Bjorntorp PA. Sex differences in the regulation of energy
balance with exercise. Am J Clin Nutr 1989; 49: 958±961.
42 Lechtig A, Yarborough C, Delgado H, Habicht J-P, Martorell
R, Klein RE. In¯uence of maternal nutrition on birth weight.
Am J Clin Nutr 1975; 28: 1223±1233.
43 van der Spuy ZM, Steer PJ, McCusker M, Steele SJ,
Jacobs HS. Outcome of pregnancy in underweight women
after spontaneous and induced ovulation. BMJ 1988; 296:
962±965.
44 Goldberg G, Prentice AM, Coward WA, Davies HL, Murgatroyd PR, Sawyer MB, Ashford J, Black AE. Longitudinal
assessment of the component of energy balance in wellnourished lactating women. Am J Clin Nutr 1991; 54: 788±
798.
45 Prentice AM, Goldberg GR, Prentice A. Body mass index and
lactation performance. Eur J Clin Nutr Suppl 1994; 48: S78±
S89.
46 Rothwell NJ, Stock MJ. A role for brown adipose tissue
in diet-induced thermogenesis. Nature Med 1979; 281: 31±
35.
47 Pond CM. Interactions between adipose tissue and the immune
system. Proc Nutr Soc 1996; 55: 111±126.
48 Stallone DD. The in¯uence of obesity and its treatment on the
immune system. Nutr Rev 1994; 52: 37±50.
49 Christensen EH, Hansen O. Arbeitsfahigkeit und ernahung.
Skand Arch Physiol 1939; 81: 160±172.
745
Bene®cial fat
NG Norgan
746
50 Romijn JA, Coyle EF, Sidossis LS, Gastadelli A, Horowitz JF,
Endert E, Wolfe RR. Regulation of endogenous fat and carbohydrate metabolism in relation to exercise intensity and duration.
Am J Physiol (Endocrinol Metab) 1993; 265: E380±E391.
51 Coyle EF. Substrate utilisation during exercise in active
people. Am J Clin Nutr 1995; 61: 968S±979S.
52 Tarnopolsky LJ, MacDougal JD, Atkinson SA, Tranopolsky
MA, Sutton JR. Gender differences in substrate for endurance
exercise. J Appl Physiol 1990; 68: 302±308.
53 Ruby BC, Robergs RA. Gender differences in substrate during
exercise. Sports Med 1994; 17: 393±410.
54 Costill DL, Fink WJ, Getchell LH, Ivy JL, Witzmann FA.
Lipid metabolism in skeletal muscle of endurance-trained
males and females. J Appl Physiol 1979; 47: 787±791.
55 Nygaard E. Energy supply and utilisation in women and
menÐa comparison. In: Vague J, Bjorntorp P, Guy-Grand
B, Rebuffe-Scrive M, Vague P (eds). Metabolic Complications
of Human Obesities. Excerpta Medica: Amsterdam, 1985, pp
249±258.
56 Hellstrom L, Blaak E, Hagstrom Toft E. Gender differences in
adrenergic regulation of lipid mobilisation during exercise. Int
J Sports Med 1996; 17: 439±447.
57 Hla MM, Davis JW, Ross PD, Wasnich RD, Yates AJ, Ravn
P, Hoskins DJ, McClung MR, EPIC Study Group. A multicenter study of the in¯uence of fat and lean mass on bone
58
59
60
61
62
63
64
mineral content: evidence for differences in their relative
in¯uence at major fracture sites. Am J Clin Nutr 1996; 64:
354±360.
Compston JE, Bhambhani M, Laskey MA, Murphy S, Khaw
KT. Body composition and bone mass in post-menopausal
women Clin Endocrinol 1992; 37: 426±431.
Sowers MR, Kshirsagar A, Crutch®eld MM, Updike S. Joint
in¯uence of fat and lean body composition compartments on
femoral bone density in premenopausal women. Am J Epidemiol 1992; 136: 257±265.
Edelstein SL, Barrett Connor E. Relation between body size
and bone mineral density in elderly men and women. Amer J
Epidemiol 1993; 138: 160±169.
Reid IR, Ames R, Evans MC, Sharpe S, Gamble G, France JT,
Lim TTM, Cundy TF. Determinants of total body and regional
bone mineral density in normal postmenopausal womenÐA
key role for fat mass. J Clin Endocrinol Metabolism 1992; 75:
45±51.
Felson DT. Weight and osteoarthritis. Am J Clin Nutr 1996;
63: 430S±432S.
Lauritzen JB. Hip fractures: incidence, risk factors, energy
absorption and prevention. Bone 1996; 18: 65S±75S.
Nesse RM, Williams GC. Evolution and Healing: The New
Science of Darwinian Medicine. Weidenfeld & Nicolson:
London, 1995.